Alibaba (Qwen) · language model
Qwen2.5 72B Instruct: pricing, context and benchmarks
Qwen2.5 72B Instruct is Alibaba (Qwen)'s model released on September 19, 2024. It costs $0.36 per million input tokens and $0.4 per million output tokens, with a 32.8K-token context window. No independent score has been published for it so far.
| Price | 20/30 | |
|---|---|---|
| Context | 1/15 | |
| Features | 12/15 |
What you need to know about Qwen2.5 72B Instruct
Alibaba (Qwen), based in China, put Qwen2.5 72B Instruct online on September 19, 2024. It accepts text as input and produces text. Its context window reaches 32,768 tokens, larger than 8 % of the models in the registry, and it can write up to 16,384 tokens in one go.
Released more than 2 years ago, it belongs to a generation most vendors have since superseded.
Qwen2.5 72B Instruct pricing
On OpenRouter, Qwen2.5 72B Instruct costs $0.36 per million input tokens and $0.4 per million output tokens. Counting three tokens read for every token written, its average price is $0.37 per million, more expensive than 32 % of paid models (2.1 times cheaper than the median of $0.76).
A concrete example: a task that sends 2,000 tokens (about 1,500 words) and receives a 500-token answer costs $0.0009. At 1,000 tasks a day, budget $28 a month. The price calculator redoes this with your own volumes.
Performance and benchmarks
Epoch AI does not yet publish a capabilities index for Qwen2.5 72B Instruct. We therefore show no made-up performance rating: only price, context and features count towards its score.
Scores published by Epoch AI on the benchmarks the site tracks:
| Benchmark | What it measures | Score | Date |
|---|---|---|---|
| GPQA Diamond | PhD-level science questions | 49.2 % | Sep 19, 2024 |
| OTIS Mock AIME | maths competition problems | 8.1 % | Sep 19, 2024 |
Features, privacy and local use
- It supports tool calling, so it can drive an AI agent, call APIs or MCP servers.
- It can answer in an enforced JSON format (structured outputs).
- Its knowledge stops at June 2024.
Its weights are open and downloadable from Hugging Face (other licence): you can run it on your own machines without sending your data to a third party. With 72,706,203,648 billion parameters, it needs about 40,627,757,263 GB of memory at 4-bit quantisation. Check with the VRAM calculator. It was downloaded 221,445 times on Hugging Face last month.
Other Alibaba (Qwen) models
| Model | Released | Avg. price | ECI |
|---|---|---|---|
| Qwen3.8 Max Prime | Sep 23, 2026 | $6 | — |
| Qwen3.8 Omni Flash | Sep 21, 2026 | $0.23 | — |
| Qwen3.8 Max (0902) | Sep 3, 2026 | $3 | 155.1 |
| Qwen3.8 Flash | Aug 26, 2026 | $0.23 | — |
| Qwen3.8 27B | Aug 14, 2026 | $0.78 | 151.7 |
| Qwen3.8 2.4T A95B | Aug 12, 2026 | $3 | — |
| Qwen3.7 Flash | Jul 27, 2026 | $0.06 | 144.1 |
| Qwen3.7 Plus | Jun 3, 2026 | $0.56 | 147 |
Verdict
Computed score: 55/100, from public data (performance, price, context, features).
Strengths
- Open weights: self-hostable, your data stays with you.
- Agent-ready: tool calling and structured JSON output.
Limitations
- No independent score published: performance cannot be verified.
- Older generation, often replaced by a newer model from the same vendor.
- Text only: reads neither images nor files.
Avoid if…
- you analyse long documents in one go;
- you want to host it on a single consumer computer (≈ 40,627,757,263 GB needed at 4-bit).
Alternatives
- Saba$0.3
- Mercury 2$0.38
- Uncensored$0.38
- Mistral Small 3.1 24B$0.4
- ReMM SLERP 13B$0.43
- Perceptron Mk1.5$0.49
Comparisons and tools
Learn more
Frequently asked questions
How much does Qwen2.5 72B Instruct cost?
$0.36 per million input tokens and $0.4 per million output tokens on OpenRouter (checked October 25, 2026). A task with 2,000 tokens read and 500 written costs $0.0009.
How big is Qwen2.5 72B Instruct's context window?
32,768 tokens, roughly 25,000 words, with a maximum of 16,384 output tokens.
Can Qwen2.5 72B Instruct run locally?
Yes, its weights are published under the other licence. Expect about 40,627,757,263 GB of memory at 4-bit.
Is Qwen2.5 72B Instruct suitable for an AI agent?
Technically yes: it supports tool calling and structured outputs. Then check its cost for your volume.
What are the alternatives to Qwen2.5 72B Instruct?
The closest models on price and performance are Saba, Mercury 2, and Uncensored.
Related articles
- Mistral AI: models, pricing and agentsOctober 25, 2026
- LLM benchmarks: how to read model rankingsOctober 24, 2026
- Running a local LLM: hardware, software, and limitationsOctober 24, 2026
- Open-source or closed models for your AI agents?October 5, 2026
Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score